Dear Nam,

My understanding is that the trajectory as printed in VASP’s XDATCAR is folded 
or wrapped. As you can see in msd.f90,:

write(6,*) 'Name of unwrapped xyz file’

So the input file should be unwrapped. In other words, you have to first unwrap 
the coordinates in XDATCAR. Then you can use msd.f90.

To unwrap the coordinates, there are other codes on that website such as 
unwrap_PBC.f90.

Cheers,

Vahid


On Oct 30, 2017, at 9:25 PM, Nam Tran 
<[email protected]<mailto:[email protected]>> wrote:

Dear Vahid

I don't know whether VASP folds back atoms in their output or not. If they do 
then the msd.f90 you recommended is not accurate. Because the code simply 
calculate the distance between two position

dRx = Rx(j) - Rx0(j)
dRy = Ry(j) - Ry0(j)
dRz = Rz(j) - Rz0(j)
sum_msd = sum_msd + (dRx**2 + dRy**2 + dRz**2)

One thing is that I tried to create a .axsf file using cppx. And the atoms seem 
to be not folded back. Can someone confirm me that?

Best Regards,
Nam Tran


________________________________
From: [email protected]<mailto:[email protected]> 
<[email protected]<mailto:[email protected]>> on behalf of 
Vahid Askarpour <[email protected]<mailto:[email protected]>>
Sent: Monday, October 30, 2017 11:00:01 PM
To: PWSCF Forum
Subject: Re: [Pw_forum] How to calculate Mean Square Displacement (MSD)

There is a collection of fortran codes (for VASP mainly but can easily be 
adapted) for processing simulation output including MSD calculation and the 
treatment of periodic images. They can be found at https://github.com/boates 
under the Physics repository (see msd.f90).

Cheers,

Vahid

Vahid Askarpour
Department of Physics and Atmospheric Science
Dalhousie University,
Halifax, NS, Canada


On Oct 30, 2017, at 4:29 AM, Nam Tran 
<[email protected]<mailto:[email protected]>> wrote:

Dear QE experts,

I am trying to evaluate MSD for my simulation.
My problem is that when an atom exits the simulation cell.
I do not know whether I should consider the distance with atom that goes about 
outside the box (a) or its image (b). Please check the image below (Blue is the 
original position t=0, red is the position at time t)

<Screenshot from 2017-10-30 14-53-57.png>
Best Regards
Nam Tran
_______________________________________________
Pw_forum mailing list
[email protected]<mailto:[email protected]>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
[email protected]<mailto:[email protected]>
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
[email protected]
http://pwscf.org/mailman/listinfo/pw_forum

Reply via email to